Request for help meshing and and analysing a part {Solved}

[johnwang]

Also I noticed the numbering matrix of the nodes eg 1, 2, 3, & 4 “reads” 4, 3, 1, 2.

1,2,5,6 is the wedge and the Array may started along Y and then spread along X.

It would be better the BooleanFragments object could sends all information into FemMsh object, so no need to use gmsh for 1D element.

[johnwang]

Fixed constrains

msh4.jpg (75.83 KiB) Viewed 20854 times

[Laurie Hartley]

Fixed constrains

Ok John - your screenshot as prompted a question in me that I should have asked, and will now ask Thomas :-

Why are there 4 different types of fixed constraints in the model?

@bernd this question is proof that you are not the greenhorn

[johnwang]

Why are there 4 different types of fixed constraints in the model?

I know for a 2D truss, 0ne end is fixed pin, the other end is a roller.
For this 3D case,:
node 6 is all fixed.
node 30 is allowed to moved along x
node 8 is allowed to move along y

What connector is used in the field to fix Z and free X,Y?

Are you good at applying wind force?

msh5.jpg (63.13 KiB) Viewed 20817 times

[Laurie Hartley]

What connector is used in the field to fix Z and free X,Y?

I have a couple of ideas on that subject which I will post the detail of tomorrow - I am on an iPad today

Are you good at applying wind force?

Short answer is No but from what I remember I think the force has to be applied to the total surface area of the exposed members which in this model are the chords, struts, and nodes - the side view?? There may also be uplift.

Maybe this can be calculated from the python console or FC Info.

Over to the experts!

EDIT - Thomas has done the wind load calculations already. See his post on Sunday August 29th at 8:58 on this thread.
Download the PDF file attached to that post. It includes the detailed calculations for wind loads!!!

[cad1234]

About Reaction Forces / Support Forces..as it was asked above...

- after the static-calculation is finished, you can open ccx_dat_file , to see the reaction forces.
- keep in mind, that CONSTRAINTDISPLACEMENT003 (ADZ) in this case covers 5 structure fixings, see FC-File.
4 of them are the inner supports, which carry 20..22 KN (z) each.. as far as I see..

Example comparison for the sum of these 5 fixings (reaction force fz):

FC→ 95 KN
MASTAN → 96,7 KN

BTW:
if you want to seperate these results, you have to set the constraint displacement for these nodes separatley, one by one.

[thschrader]

Just for fun and to have another comparision I did the same buckling calculation in Code_Aster (I'm still learning this software).
Here are my results which BTW are in great accordance with what already was found:

Thanks for checking!
Is there a tool/command in aster to add hinges at the nodes?
I did this with Mastan (lot of work to get a stable structure).
When zooming to a node, it seems the hinges are simulated by adding
additional elements (with reduced stiffness?). You can do the same with FC.

beam2field.FCStd

(134.43 KiB) Downloaded 409 times

[Laurie Hartley]

@cad1234 - I am trying to understand this - I will try to explain.

JohnWang has numbered the nodes above. There are 2 rows of 4 supporting nodes. The inner supporting nodes are 5,7,29,&31. Based on the symmetry of the structure and the uniform distribution of nodal loads, I would expect the reactions on each of these nodes to be twice that on the outer supporting nodes 6,8,30,&32.

I am still on an iPad so I am not sure but I thought the total reaction would have been closer to 120kn based on Thomas’ inclusion of self weight and ice loads.

Nevertheless, if the total reaction is indeed approximately 96kn / ((4x2)+4) = 8kn. Therefore “inner supporting nodes” support 16kn each and “outer supporting nodes” support 8kn each. Am I being too simplistic?

[cad1234]

@thomas
I checked analysis type “space truss” for hinge simulation, which runs well here (without additional work),
no momentum from the analysis, max disp: 1.89 mm, max element force: -21.4 KN
There is only one requirement:

No subs (the same in Mecway), as the additional nodes would be interpreted as hinge too..

I guess (not sure about) that we do not need subs, if we have only normal forces..
buckling does not work this way, as it needs subs..
What do you think about?
Maybe I´m completely wrong..

[thschrader]

I guess (not sure about) that we do not need subs, if we have only normal forces..
buckling does not work this way, as it needs subs..
What do you think about?

Static analysis:
As long as the forces are acting on the nodes, you dont need hinges. For a structure like this.
You can generate equilibrium with normal-forces alone, no bending moments needed.

Buckling:
That is another point. See pics: with hinges you get different buckling-modes and different
load-factors. The beam at the top with the hinges buckles "alone".

That means:
For stress calculation run first order analysis.
For checking global system stability run geometric nonlinear analysis.
Check local beam stability (Euler).

spaceframe_buckling_hinges.FCStd

(273.76 KiB) Downloaded 438 times

hinges002.JPG (44.17 KiB) Viewed 20529 times